1
MuCon: Clipped Muon Updates for LLM Training
MuCon通过裁剪奇异值改进Muon优化器，为LLM训练提供更稳定的更新策略
arXiv:2605.26459v1 Announce Type: new Abstract: Muon-style optimizers take a matrix-valued momentum or preconditioned update $B = U \operatorname{diag…
MuCon通过裁剪奇异值改进Muon优化器，为LLM训练提供更稳定的更新策略
arXiv:2605.26459v1 Announce Type: new Abstract: Muon-style optimizers take a matrix-valued momentum or preconditioned update $B = U \operatorname{diag…